Review of the Methodological Approaches to the Study of the Impact of Microenterprise Credit Programs
Covers eleven studies carried out in Asia, Africa and Latin America. Focuses on sample design and execution, temporal issues, analytical technique and control methods for eliminating other explanations for changes. Explains issues of selection bias and fungibility, meaning that financial and other resources including credit/profits from microenterprise projects may move between and among various household activities making it difficult to track impacts.
Recommendations include:
- Sampling for assessment should occur within control groups from programme sites and a control population from matched non-programme sites;
- Statistically equated control groups may be used for individual controls;
- In programme sites, eligible non- borrowers should be used for individual controls;
- Sample size of 500 allows for effective use of control variables and for dealing with problems of longitudinal analysis.
Suggests that assessment should give consideration to time and longitudinal impact, and recommends 18-24 month intervals for data collection since the point at which impacts first begin to occur and the length of time that impacts are sustained are subjects of debate.
